Design And Implementation Of Personalized Recommender System For Smart Trading Area

With the rapid development of Internet of things,and mobile Internet technology,tradition trading area is gradually replaced by smart trading area.Smart trading area,as a part of smart city,relying on the line under the 020 interactive mode,has been a new way to promote consumption.Therefore,how to quickly locate the line of real goods for the majority of online consumers' needs in order to promote the final consumption,become a key issue in the construction of smart trading area.To solve the above problems,one popular solution is recommender system.With the increasing amount of data in the smart trading area,how to find the user's favorite products in the mass of data has been very important.Therefore,based on scenarios of smart trading area,this paper analyzes potential needs,and formulate a design plan of the distributed architecture based on a real-time system,and then implement a real-time recommendation system,which can be applied to smart trading area,to quickly provide personalized recommendation for the needs and interest of users.In order to achieve the above goals,the main work of this paper is as follows:1.In order to meet the real-time requirements of the recommender system,this paper studies on big data open source framework,and then designs and implements an offline-online-realtime tier architecture based on the scene of smart trading area.2.Based on current incremental recommender model,this paper propose an incremental recommender model which is suitable for smart trading area.3.Based on the advantages and weaknesses of various types of recommendation algorithm,this paper proposes the most suitable offline and online recommendation algorithm with high accuracy,high performance based on the scene of smart trading area.4.Combined with the scene of smart trading area,data source in smart trading area system is worked as a system of the recommended system configuration,and our recommender system supports a web service to smart trading area system,and finally achieve an integrated solution of recommender system with smart trading area system.5.The accuracy,real-time performance of real-time recommendation system are designed to test in large data scale.The final experiment and system verification proved that the design and implementation of real-time recommendation system provides a high accurate real-time solution for smart trading area,and even more application scenarios,which undoubtedly has a certain theoretical value and application prospects.